Cloudflare unveils advanced AI-Powered security solutions

Cloudflare revealed a range of novel security products on Monday that utilize artificial intelligence (AI) to safeguard against AI-driven threats, secure AI solutions, and counteract AI-enhanced phishing.

To enhance user analytics capabilities, Cloudflare introduced an AI assistant within the Security Analytics segment of its dashboard. This AI assistant aids in swift identification of anomalies and cyberattacks by simplifying the process of navigating multiple tools and filters. Employing Cloudflare's Workers AI, the assistant generates charts based on user queries, applying filters and consolidating the necessary data and logs into a single graph.

The new AI assistant enables customers to use natural language to gain insights into their application security, including detecting spikes in traffic, determining the root cause of 5xx errors, identifying bot attacks and unwanted automated traffic, among other functionalities. Currently available in beta for business and enterprise customers, the solution supports basic inquiries.

Cloudflare employs Defensive AI, a framework utilizing data from its network to enhance protection against various threats, including those powered by AI. This approach extends across multiple security areas, from application security to email security and the Zero Trust platform. Cloudflare emphasizes customizing protection for each customer, whether for API or email security, and leveraging vast amounts of attack data to train models for detecting undiscovered application attacks.

To safeguard APIs, Cloudflare is developing API Anomaly Detection for API Gateway, which utilizes machine learning to understand the business logic of applications and create a model for identifying attacks. Customers can sign up for the beta release of API Anomaly Detection. Various Cloudflare products, such as Web Application Firewall (WAF), Email Security service, and Zero Trust, rely on AI to identify cyberattacks, analyze phishing patterns, and detect anomalies in user behavior for user risk scoring.

In light of the increased use of AI in cyberattacks, particularly in personalized phishing lures, Cloudflare remains confident in its preemptive scanning approach to detecting malicious emails. Despite evolving attacker techniques, Cloudflare asserts its ability to defend customers against emerging threats.

Cloudflare also announced Firewall for AI on Monday, designed to protect Large Language Models (LLMs) by detecting vulnerabilities, analyzing user-submitted prompts to identify potential abuse attempts, preventing denial of service attempts, and identifying sensitive information returned by the model. Firewall for AI incorporates tools from WAF, such as Rate Limiting and Sensitive Data Detection, along with a prompt validation feature for identifying abuses like prompt injection attempts, currently in development. Advanced Rate Limiting and Sensitive Data Detection are already available for enterprise customers, while the prompt validation feature will be released for all Workers AI users in the coming months.